STOKA / Systems

The software that makes an automated warehouse work

Software is what decides. An automated warehouse without software is a set of machines that does not know what to move or in what order. The software defines where each item is stored, in what priority it is retrieved and which machine does it, and records every movement. STOKA designs, imports, installs, integrates and supports these systems, and connects them to what your company already runs. This page explains each layer in plain language before its acronym, clarifies the difference with your business management system and tells you when you do not need it yet.

Tell us your problem

Why does the team lose time looking for stock the system says is there?

Because your company system knows how much stock there is, but not where it sits. If the location is written by hand, or lives in the head of whoever picked the last order, every search turns into a walk around the warehouse. Warehouse software records the exact position on every movement and turns that walk into an instruction.

What is warehouse software and what does each layer do?

In plain language: it is the part that thinks. There are four layers stacked on top of each other. The first decides what to move and where to: it holds the inventory, assigns locations, sequences orders and triggers replenishment. That layer is the warehouse management system (WMS). The second turns that decision into concrete orders for each piece of equipment and coordinates who moves first, so that two machines do not claim the same aisle at once: that is the warehouse control system (WCS). The third handles the fine detail of the handling equipment, the movement that picks the box up and sets it down where it belongs: that is the handling management system (HMS). The fourth is three-dimensional visualisation, which shows the state of the warehouse on screen and lets you see where the bottleneck is forming before it stops everything. STOKA implements all four layers and integrates them with what you already run.

WMS against ERP: the question that comes up in every meeting

It is the most frequent doubt, worth clearing up early. The business management system (ERP, enterprise resource planning) is the one the company already has: it invoices, purchases, keeps the accounts and knows how much stock exists per reference.

Keep reading

The warehouse management system knows something else: where each unit sits, in what condition, from which batch, with what expiry and in what order to retrieve it. The ERP answers how much do I have; the WMS answers where is it and what to pick first. That is why an ERP does not replace a WMS: it can hold an impeccable accounting balance and still not tell anyone which rack to take the pallet from. A WMS does not invoice or purchase either. They should talk to each other. The ERP sends the order, the WMS resolves execution inside the warehouse and returns the confirmed movement with its location. If you run an ERP with a simple location module and few daily movements, that may be enough: a WMS earns its place when location, sequence and traceability cost hours every day.

From the order to the machine: control and handling

Between the decision and the physical movement there is a translation. The warehouse management system says move the pallet from position A to the dispatch bay; the control system turns that sentence into orders for each piece of equipment, in the right sequence and with the right permissions: which conveyor starts, which machine enters the aisle, which one waits.

Keep reading

Without that layer, two machines can claim the same stretch at the same time and the warehouse stops on its own. The control system also administers priorities: an urgent dispatch goes ahead of a replenishment, and a replenishment goes ahead of an internal reorganisation. The handling layer is finer still: it manages the detail of the equipment that picks and deposits goods, the grip sequence, the exact support point and the confirmation that the unit ended up where it was meant to. The more automated the warehouse, the more this translation weighs, because it is what stops installed capacity from sitting idle waiting for an order.

Seeing the warehouse on screen before it grinds to a halt

Three-dimensional visualisation is not an animation for the boardroom. It is the real state of the warehouse rendered on screen: which positions are occupied, which equipment is working, which orders are queued and where pending work is piling up.

Keep reading

It is useful for two concrete things. The first is finding the bottleneck before it stops everything: when the queue always grows at the same point, the problem is not the machine that stopped but the stretch before it, and a spreadsheet does not show that. The second is operating without walking the warehouse to understand what is going on, which matters when the system works at height or in a cold room, where direct visual inspection is uncomfortable or simply not viable. It also earns its place during commissioning, to compare real behaviour against what the project sized and correct parameters before they settle into habit.

Integration and accurate inventory: talking to what you already run

Warehouse software does not replace the company systems: it connects to them. The connection is made through an application programming interface (API), the standard way for two systems to exchange data without anyone retyping it: the business management system sends orders and purchase orders, and the warehouse system returns goods-in confirmations, locations and movements.

Keep reading

The scope of that integration is defined against the system you already run and against the interface it exposes; it is not promised before we have seen it. The most measurable benefit shows up in inventory: with every movement recorded automatically, accuracy stops depending on a spreadsheet filled in afterwards, and stocktaking stops being a full day of people. That time has a price: a warehouse operative represents USD 965 per month, USD 11,580 per year in company cost (CCT 40/89 logistics branch, June 2026), with social charges at 34.6% of labour cost (IARAF, February 2026).

What defines the scope of the software you need?

We do not publish model-by-model specification tables: the detailed sheet comes with the assessment, sized against your operation. These are the criteria the choice is made on.

Whether locations are already assigned
This is the first filter and the most ignored. If goods are stored wherever there is room rather than in an identified position, the software has nothing to work on. Sorting locations out is process work and it comes before any licence.
Movements per day and number of references
Few movements and few references are sustained by the location module of your business management system. When movements are many, repetitive and sequenced by batch or expiry date, a dedicated warehouse management system earns its place.
What automated equipment exists, or will exist
If the warehouse is manual, the conversation starts and ends with the warehouse management system. Every automated machine added brings in the control layer, and handling equipment brings in the layer that manages its detail as well.
What interface your current system exposes
Integration runs through an application programming interface, and not every system exposes the same one. The real scope of the exchange depends on it: which data travels in each direction and how often it synchronises. It is verified before quoting.
Traceability required by batch, expiry or serial number
Food, pharmaceuticals and anything with an expiry date force you to follow the unit, not just the balance. That requirement defines the data structure of the system and is often what tips the decision, more than the number of pallets.

Engineering for your operation

When do you not need to buy software yet?

Our approach

If you have not assigned locations or standardised the load unit yet, no software will give you reliable data. You would be asking it to sort out a mess it does not know: the system will accurately record positions that do not exist on the floor, and the result is a tidy inventory on screen and an equally long walk in the warehouse. Process first. Clear out stock that does not move, define a location code, standardise the pallet and the box, and set who confirms each movement: that round costs little and it is what later makes the software trustworthy. You also do not need a dedicated warehouse management system if the warehouse is small, with few references and few movements per day, and the business management system you already run has a location module that copes. And if the problem is travel distance, software will not fix it: that is a layout job.

Frequently asked questions

Is the business management system I already have not enough?

It may well be, and in small warehouses it usually is. The business management system (ERP) knows how much stock there is; the warehouse management system (WMS) knows where it sits, in what condition and in what order it should be retrieved. If those three questions are answered today by searching, the ERP module has fallen short. If they answer themselves, buy nothing.

Does it integrate with the system we already run?

It integrates through an application programming interface (API), the standard way for two systems to exchange data without retyping anything. The scope depends on the interface your current system exposes, so it is verified against your installation before quoting. We do not promise an integration before seeing what the other side exposes.

Can we start with the software and automate later?

Yes, and it is generally the order we recommend. The warehouse management system organises locations, sequences and data about the operation you already have, and that data is what later allows an automated system to be sized properly. Automating first and organising afterwards is the expensive route: you buy capacity against assumptions instead of measured movements.

How long does commissioning take and when does it pay back?

For automated systems, installation takes 3 to 4 months from the purchase order, and software commissioning runs in parallel with the mechanical build. Payback sits between 18 and 36 months depending on the operation (STOKA criterion). A software-only project, on a manual warehouse, is resolved in shorter timescales.

Before buying software, it pays to see how the warehouse works

The on-site survey measures movements per day, references, load unit and the state of your locations, and reviews what interface your current system exposes. That is how we define the real scope. And if what you already run is enough, we will tell you.